Retraction Notice: Enhancing Collaborative Intrusion detection networks against insider attack using supervised learning technique
Rajesh Singh, Rajesh Deorari · 2022 IEEE 2nd Mysore Sub Section International Conference (MysuruCon) · 2022
As network are being used, intrusion occurs from a variety of sources. According to their actions to damage network functioning, cyberattack is a significant issue in communication systems. The privacy, authenticity, or accessibility of a resource may be compromised by unauthorized activity or attacks, which intrusion detection systems (IDS) can help identify. Researchers explore the findings of research articles on the past, present, and potential of intrusion detection systems on wireless networks. Statistical modelling and deep learning are used to analyses the findings of journals. Isolation forest are one of the strategies that can acquire accurate network measures and produce appropriate choices by raising the detection rate and accuracy. In this paper, we propose a decision tree-based method for detection of network intrusions with improved data quality. In order to improve the data quality and provide appropriate training, networking data pre-processing and attributes selection based on entropy are specifically carried out. A isolation forest classifier is then constructed for accurate intrusion detection. Experimental analysis of two datasets demonstrates the proposed model's ability to produce reliable results. Actually, using the CICIDS2017 and NSL-KDD datasets, our model's accuracy is 99.98% and 99.82%, respectively. When compared to existing models, the new approach has many benefits in terms of false alarm rate (FAR), detection rate (DR), and accuracy (ACC).